Complementary metal-oxide semiconductor (CMOS) image sensors are gaining in popularity over traditional charged-coupled devices (CCDs) due to certain advantages inherent in the CMOS image sensors. In particular, CMOS image sensors typically require lower voltages, consume less power, enable random access to image data, may be fabricated with compatible CMOS processes, and enable integrated single-chip cameras. Generally, CMOS image sensors utilize light-sensitive CMOS circuitry to convert light energy into electrical energy. The light-sensitive CMOS circuitry typically includes a photo-diode formed in a silicon substrate. As the photo-diode is exposed to light, an electrical charge is induced in the photo-diode. The photo-diode is typically coupled to a MOS switching transistor, which is used to sample the charge of the photo-diode. Colors may be determined by placing filters over the light-sensitive CMOS circuitry.
Typically, CMOS image sensors are fabricated utilizing a capacitance within the photo-diode and a floating capacitance created between transistor connections. These capacitances, however, are characterized by small capacitance values, which cause a high susceptibility to noise and reduce the maximum output signal. Attempts have been made to increase the signal generated by a photo-diode, but these increase the electrical charge generated by the photo-diode and do not necessarily increase the output signal. Furthermore, CMOS image sensors are typically fabricated utilizing MOS transistors having a polysilicon gate and silicon nitride spacers. This type of transistor, however, introduces a silicon surface trap and leakage. As a result, noise on the output signal increases and the dark signal increases.
The CMOS image sensors typically need one or more capacitors per pixel array. In the conventional frontside illuminated (FSI) image sensor application, the photodiode region, the transistor region, and the capacitor region are implemented on one wafer during the same fabrication process, and the distance from a lens to the photodiode becomes large due to a back-end-of-the-line (BEOL) metal line formed on the transistor region and the capacitor region, thus there is a great loss of the light signal entered into the photodiode region. The transistor region and the capacitor region coexist in one image sensor device, which results in a region where all of the light incident cannot be absorbed but, rather, is lost.